CN105163204B - A kind of real-time slice systems of files in stream media and method - Google Patents
A kind of real-time slice systems of files in stream media and method Download PDFInfo
- Publication number
- CN105163204B CN105163204B CN201510540681.5A CN201510540681A CN105163204B CN 105163204 B CN105163204 B CN 105163204B CN 201510540681 A CN201510540681 A CN 201510540681A CN 105163204 B CN105163204 B CN 105163204B
- Authority
- CN
- China
- Prior art keywords
- data
- program
- pcie
- slice
- boards
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/80—Generation or processing of content or additional data by content creator independently of the distribution process; Content per se
- H04N21/83—Generation or processing of protective or descriptive data associated with content; Content structuring
- H04N21/845—Structuring of content, e.g. decomposing content into time segments
- H04N21/8456—Structuring of content, e.g. decomposing content into time segments by decomposing the content in the time domain, e.g. in time segments
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
- H04N21/433—Content storage operation, e.g. storage operation in response to a pause request, caching operations
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Television Signal Processing For Recording (AREA)
Abstract
The invention discloses a kind of real-time slice systems of files in stream media and method, including:Content server, the configuration information that network video server is sent is received, is configured accordingly according to configuration information so that the program data of output corresponds with network port number;Input signal is received simultaneously, input signal is handled, and the program data that processing is obtained is sent to PCIe boards;PCIe boards, the program data that content server is sent is received, the program data received is cut into slices in real time, the slice of data for obtaining program is stored in the memory chip of PCIe boards;Network video server, provided with disk array, corresponding configuration information is sent to content server, and program slice of data is read from PCIe boards by the PCIe interface of PCIe boards, judge whether program slice of data is effective, if effectively, storage slice of data to disk array.Have the advantages that good concurrency, processing section fast response time, stability are high.
Description
Technical field
The present invention relates to streaming media playing field, more particularly to a kind of real-time slice systems of files in stream media and method.
Background technology
With the fast development of network technology, various video websites emerge in an endless stream, and either internet video live broadcasting, program request be also
Time shift, the fluency and requirement of real-time more and more higher that streaming media plays, the precision of corresponding streaming media section and
Requirement of real-time also more and more higher.Therefore, fast and accurately streaming media is cut into slices and transmitted and just become to video website
Basic demand.
But the following problem of network video server generally existing of existing video website:
1st, concurrency is poor.Existing network video server is all based on windows software platforms, playing stream media file
When, it is required for an independent thread to be handled per programs, but due to processor (CPU) limitation so that receive line
Journey is that usually order is carried out, even if multi-thread concurrent can be supported to work in view of processor, is capable of the thread of concurrent efforts
Quantity be also conditional.
2nd, delay and packet loss are serious.It is that order performs for CPU due to receiving program data by software, therefore
CPU can only could receive other programs after a programs have been received, when there is programs up to a hundred to need to receive, Suo Youjie
Mesh all receives the time at least needing up to a hundred milliseconds, can cause programme televised live playout-delay, and program data packet loss
Problem.
3rd, system resource takes big.During receiving, program data (or the program of certain length can be cached in internal memory
Section), in general, the bandwidth of SD program is generally 6-8Mbps, if the program data of caching 2s durations, often set saves
The spatial cache that mesh needs is about 1.5MBytes (6Mbps*2s/8) -2MBytes (8Mbps*2s/8), then 100 programs need
The spatial cache wanted is then 150MB-200MB, takes the substantial amounts of memory headroom of server.In addition, number of programs is realized by software
It is to CPU requirement and very high according to reception, it is necessary to take substantial amounts of cpu resource always.
The content of the invention
The purpose of the present invention is exactly to solve the above problems, there is provided a kind of real-time slice systems of files in stream media and side
Method, it has the advantages that good concurrency, processing section response quickly, stability are high.
To achieve these goals, the present invention adopts the following technical scheme that:
A kind of real-time slice systems of files in stream media, including:
Content server, it is attached by network and network video server, receives what network video server was sent
Configuration information, configured accordingly according to configuration information so that a pair of the program data of output and network port number port 1
Should;Input signal is received simultaneously, input signal is handled, and the program data that processing is obtained is sent to PCIe boards;
PCIe boards, it is connected by network with the content server, receives the program data that content server is sent, it is right
The program data received is cut into slices in real time, and obtained program slice of data is stored in the memory chip of PCIe boards
In;
Network video server, provided with disk array, corresponding configuration information is sent to content server, and pass through PCIe
The PCIe interface of board reads program slice of data from PCIe boards, judges whether program slice of data is effective, if effectively,
Program slice of data is stored to disk array, otherwise, is not stored.
The PCIe boards include fpga chip, fpga chip respectively with FLASH chip, memory chip, network interface and
PCIe interface is bi-directionally connected;Wherein, fpga chip is cut into slices in real time to receiving program data, and the slice of data that will be obtained
It is stored in memory chip.
The memory chip is DRR3.
It is provided with PCIe inside the fpga chip to cache, for caching the slice of data read from memory chip.
A kind of real-time dicing method of files in stream media, comprises the following steps:
S1:Network video server sends program configuration information to content server;
S2:Content server receives program configuration information and configured according to program configuration information, receives input signal;
S3:Content server parses to input signal, obtains per programs data, will be sealed respectively per programs data
Network packet is dressed up, and the network packet of every programs is sent to PCIe boards by the different network ports respectively;
S4:PCIe boards receive the network packet that content server is sent and parsing, obtain per programs data;
S5:PCIe boards are cut into slices to every programs data, and program slice of data is stored in PCIe boards
Deposit in chip;
S6:Network video server reads the program slice of data of program from PCIe boards and is stored in disk array
In.
The network packet is UDP message bag.
The PCIe boards are concurrently cut into slices in real time using the system clock of fpga chip to every programs data.
Also include before the step S4:
S4-0:PCIe boards obtain section from the FLASH chip of the network video server or PCIe boards and matched somebody with somebody
Confidence ceases, and carries out corresponding configuration according to section configuration information.
The step S6 is further:
S6-1:Network video server sends reading data command and gives PCIe boards;
S6-2:The fpga chip of PCIe boards reads program slice of data from memory chip, and is stored in PCIe cachings
In;
S6-3:Network video server reads program slice of data from the PCIe cachings of PCIe boards, and judges program
Whether slice of data is valid data;If valid data, then step S6-4 is performed;Otherwise, if not valid data, then prompt
Error message;
S6-4:By the storage of program slice of data into disk array, go to step S6-2 and continue executing with.
The step S6 further comprises:Network video server judges whether to receive pause section order, if receiving
To pause section order, then sent to PCIe boards and stop section order, PCIe boards stop section.
Beneficial effects of the present invention:
The present invention uses PCIe board hardware, based on the system clock of PCIe boards, more set single-unit mesh streams is carried out concurrent
Reception and section, both solved the shortcomings that prior art concurrency difference, also compensate for delay and packet loss it is serious the defects of.Separately
Outside, because section work is realized by PCIe boards hardware, it is not necessary to which network video server participates in, releasing network video clothes
The substantial amounts of memory headroom of business device and cpu resource, greatly reduce network video server pressure.
Brief description of the drawings
Fig. 1 is the real-time slice systems structural representation of files in stream media of the present invention;
Fig. 2 is PCIe board structures schematic diagram of the present invention;
Fig. 3 is the real-time dicing method schematic flow sheet of files in stream media of the present invention;
Fig. 4 is that inventive network video server reads program slice of data flow chart from PCIe boards;
Fig. 5 is storage schematic diagram of the program slice of data of the present invention in memory chip.
Embodiment
The invention will be further described with embodiment below in conjunction with the accompanying drawings.
As shown in figure 1, a kind of real-time slice systems of files in stream media, including:
Content server, it is attached by network and network video server, receives what network video server was sent
Configuration information, configured accordingly according to configuration information so that the single program stream and network port number (port) of output are one by one
It is corresponding;Input signal is received simultaneously, input signal is handled, and single program stream that processing is obtained is sent to PCIe plates
Card;
PCIe boards, it is connected by network with the content server, receives the program data that content server is sent, it is right
The program data received is cut into slices in real time, and obtained slice of data is stored in memory chip;
Network video server, provided with disk array, corresponding configuration information is sent to content server, and pass through PCIe
Interface reads program slice of data from PCIe boards, judges whether slice of data is effective, if effectively, storage slice of data arrives
Disk array, otherwise, do not store.
The content server is to be connect with ASI (Asynchronous Serial Interface, Asynchronous Serial Interface)
The equipment of mouth, ASI input signals are received, input signal is demultiplexed, obtain program data, and program data is packaged into
Network packet, PCIe boards are then sent to by the network port number of setting.
The content server is PC, reads program data from internal disk, is then packaged into program data
Network packet, PCIe boards are then sent to by the network port number of setting.Program data in the internal disk is
PC is obtained and is stored in advance in disk from external input signal.
As shown in Fig. 2 PCIe boards include fpga chip, fpga chip is respectively at FLASH chip, memory chip, network
Interface and PCIe interface are bi-directionally connected;Wherein, fpga chip is cut into slices in real time to receiving program data, and is cut what is obtained
Sheet data is stored in memory chip.
The memory chip is DRR3, for storing program slice of data.Memory chip is the important component of PCIe boards,
Using DDR3, it is very fast to be primarily due to DDR3 read or write speeds.DDR3 chips mainly have at present DDR3-800, DDR3-1066,
The models such as DDR3-1333, DDR3-1600, by taking the DDR3-800 of minimum frequency as an example, its Peak bit rate is 6400MB/s.Often
The section duration of programs calculates according to 2s, and the slice size of a set of SD program is about 2MByte, and DDR3 reads this section
Only need 312.5ns.
It is provided with PCIe inside the fpga chip to cache, for caching the slice of data read from memory chip.
The FLASH chip is stored with section configuration information, and PCIe boards pass through network interface according to section configuration information
Single program stream data is received from different network port numbers (port).
The network interface is kilomega optic fiber network interface.
As shown in figure 3, a kind of real-time dicing method of files in stream media, comprises the following steps:
S1:Network video server sends program configuration information to content server;
S2:Content server receives program configuration information and configured according to program configuration information, receives input signal;
S3:Content server parses to input signal, obtains per programs data, will be sealed respectively per programs data
Network packet is dressed up, and the network packet of every programs is sent to PCIe boards by the different network ports respectively;
S4:PCIe boards receive the network packet that content server is sent and parsing, obtain per programs data;
S5:PCIe boards are cut into slices to every programs data, and slice of data is stored in the internal memory core of PCIe boards
In piece;
S6:Network video server reads the slice of data of program from PCIe boards and is stored in disk array.
The program configuration information comprises at least per programs and network port number one-to-one relationship so that per programs
Data are sent to PCIe boards from the specified network port respectively.Such as the map network port numbers port10011 of program 1, program 2
Map network port numbers prot10012 ... ...
The network packet is UDP message bag, and network packet is sent to PCIe by content server by udp protocol
Equipment.
The PCIe boards are concurrently cut into slices in real time using FPGA system clock to every programs data.
Also include before the step S4:PCIe boards are from the network video server or the FLASH of PCIe boards
Section configuration information is obtained in chip, and corresponding configuration is carried out according to section configuration information.
Further comprise as shown in figure 4, the step S6 network video servers read slice of data from PCIe boards
Following steps:
S6-1:Network video server sends reading data command and gives PCIe boards;
S6-2:The fpga chip of PCIe boards reads program slice of data from memory chip, and the PCIe cachings deposited
In;
S6-3:Network video server reads program slice of data from the PCIe cachings of PCIe boards, and judges program
Whether slice of data is valid data;If valid data, then step S6-4 is performed;Otherwise, if not valid data, then prompt
Error message;
S6-4:By the storage of program slice of data into disk array, go to step S6-2 and continue executing with.
The step S6 further comprises:Network video server judges whether to receive pause section order, if receiving
To pause section order, then sent to PCIe boards and stop section order, PCIe boards stop section.
As shown in figure 5, when program slice of data is deposited in PCIe board memory chips, one section of section is deposited per programs
Data.Assuming that the section per programs stream is 2s, the section 2MBytes spaces of every programs, the corresponding network port are distributed to
Number postponed since port10011, then, the map network port numbers port10011 of program 1, the storage position in memory chip
It is set to 0~2MBytes;The map network port numbers port10012 of program 2, deposit position in memory chip for 2~
4Mbytes, network port number corresponding to 3~n of program and the storage location in memory chip can be so that the rest may be inferred, program n
Map network port numbers are port (10010+n), and the deposit position in memory chip is 2* (n-1)~2*n Mbytes.
Although above-mentioned the embodiment of the present invention is described with reference to accompanying drawing, model not is protected to the present invention
The limitation enclosed, one of ordinary skill in the art should be understood that on the basis of technical scheme those skilled in the art are not
Need to pay various modifications or deformation that creative work can make still within protection scope of the present invention.
Claims (9)
1. a kind of real-time slice systems of files in stream media, it is characterized in that, including:
Content server, it is attached by network and network video server, receives the configuration that network video server is sent
Information, configured accordingly according to configuration information so that the program data of output corresponds with network port number port;Together
When receive input signal, input signal is handled, and PCIe boards are sent to by obtained program data is handled;
PCIe boards, it is connected by network with the content server, the program data that content server is sent is received, to receiving
To program data cut into slices in real time, and the slice of data for obtaining program is stored in the memory chip of PCIe boards;
Network video server, provided with disk array, corresponding configuration information is sent to content server, and pass through PCIe boards
PCIe interface program slice of data is read from PCIe boards, judge whether program slice of data effective, if effectively, storing
Otherwise slice of data, does not store to disk array.
2. the real-time slice systems of files in stream media as claimed in claim 1, it is characterized in that, the PCIe boards include FPGA cores
Piece, fpga chip are bi-directionally connected with FLASH chip, memory chip, network interface and PCIe interface respectively;Wherein, fpga chip
Cut into slices in real time to receiving program data, and obtained slice of data is stored in memory chip.
3. the real-time slice systems of files in stream media as claimed in claim 2, it is characterized in that, the memory chip is DRR3.
4. a kind of real-time dicing method of files in stream media, it is characterized in that, comprise the following steps:
S1:Network video server sends program configuration information to content server;
S2:Content server receives program configuration information and configured according to program configuration information, receives input signal;
S3:Content server parses to input signal, obtains per programs data, will be packaged into respectively per programs data
Network packet, and the network packet of every programs is sent to PCIe boards by the different network ports respectively;
S4:PCIe boards receive the network packet that content server is sent and parsing, obtain per programs data;
S5:PCIe boards are cut into slices to every programs data, and program slice of data is stored in the internal memory core of PCIe boards
In piece;
S6:Network video server reads the program slice of data of program from PCIe boards and is stored in disk array.
5. the real-time dicing method of files in stream media as claimed in claim 4, it is characterized in that, the network packet is UDP numbers
According to bag.
6. the real-time dicing method of files in stream media as claimed in claim 4, it is characterized in that, the PCIe boards use FPGA systems
System clock is concurrently cut into slices in real time to every programs data.
7. the real-time dicing method of files in stream media as claimed in claim 4, it is characterized in that, also include before the step S4:
S4-0:PCIe boards obtain section from the FLASH chip of the network video server or PCIe boards and match somebody with somebody confidence
Breath, and corresponding configuration is carried out according to section configuration information.
8. the real-time dicing method of files in stream media as claimed in claim 4, it is characterized in that, the step S6 is further:
S6-1:Network video server sends reading data command and gives PCIe boards;
S6-2:The fpga chip of PCIe boards reads program slice of data from memory chip, and is stored in PCIe cachings;
S6-3:Network video server reads program slice of data from the PCIe cachings of PCIe boards, and judges that program is cut into slices
Whether data are valid data;If valid data, then step S6-4 is performed;Otherwise, if not valid data, then prompt mistake
Information;
S6-4:By the storage of program slice of data into disk array, go to step S6-2 and continue executing with.
9. the real-time dicing method of files in stream media as claimed in claim 4, it is characterized in that, the step S6 is further also wrapped
Include:Network video server judges whether to receive pause section order, if receiving pause section order, sent out to PCIe boards
Stopping section order is sent, PCIe boards stop section.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510540681.5A CN105163204B (en) | 2015-08-28 | 2015-08-28 | A kind of real-time slice systems of files in stream media and method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510540681.5A CN105163204B (en) | 2015-08-28 | 2015-08-28 | A kind of real-time slice systems of files in stream media and method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN105163204A CN105163204A (en) | 2015-12-16 |
CN105163204B true CN105163204B (en) | 2018-03-23 |
Family
ID=54803943
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510540681.5A Active CN105163204B (en) | 2015-08-28 | 2015-08-28 | A kind of real-time slice systems of files in stream media and method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105163204B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106951380A (en) * | 2017-03-17 | 2017-07-14 | 山东浪潮商用系统有限公司 | A kind of client, service end, data transmission method and system |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101931776A (en) * | 2010-06-21 | 2010-12-29 | 杭州海康威视数字技术股份有限公司 | Data storage system in video monitoring and method for storing, previewing and reviewing data |
CN102123275A (en) * | 2010-12-30 | 2011-07-13 | 北京中科大洋科技发展股份有限公司 | Video component data information acquiring and extracting method |
CN102547478A (en) * | 2012-02-20 | 2012-07-04 | 北京蓝汛通信技术有限责任公司 | Triggered slice on-demand system and method of streaming media based on CDN (Content Distribution Network) |
CN104023244A (en) * | 2014-05-29 | 2014-09-03 | 深圳市云宙多媒体技术有限公司 | Method and apparatus for slicing stream media data in CDN system |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US9483431B2 (en) * | 2013-04-17 | 2016-11-01 | Apeiron Data Systems | Method and apparatus for accessing multiple storage devices from multiple hosts without use of remote direct memory access (RDMA) |
-
2015
- 2015-08-28 CN CN201510540681.5A patent/CN105163204B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101931776A (en) * | 2010-06-21 | 2010-12-29 | 杭州海康威视数字技术股份有限公司 | Data storage system in video monitoring and method for storing, previewing and reviewing data |
CN102123275A (en) * | 2010-12-30 | 2011-07-13 | 北京中科大洋科技发展股份有限公司 | Video component data information acquiring and extracting method |
CN102547478A (en) * | 2012-02-20 | 2012-07-04 | 北京蓝汛通信技术有限责任公司 | Triggered slice on-demand system and method of streaming media based on CDN (Content Distribution Network) |
CN104023244A (en) * | 2014-05-29 | 2014-09-03 | 深圳市云宙多媒体技术有限公司 | Method and apparatus for slicing stream media data in CDN system |
Also Published As
Publication number | Publication date |
---|---|
CN105163204A (en) | 2015-12-16 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN112765054A (en) | High-speed data acquisition system and method based on FPGA | |
EP4054190A1 (en) | Video data encoding method and device, apparatus, and storage medium | |
CN107172037B (en) | Real-time subpackage analysis method for multi-path multi-channel high-speed data stream | |
US9755982B2 (en) | Method and apparatus for checking data frame length | |
WO2020001315A1 (en) | Method and apparatus for detecting video playing quality | |
FR2979164A1 (en) | CONTROLLER FOR MEMORIZATION DEVICES AND METHOD FOR CONTROLLING MEMORY DEVICES | |
US9258834B2 (en) | Method of mobile terminal internal communications | |
CN105163204B (en) | A kind of real-time slice systems of files in stream media and method | |
US20140068008A1 (en) | Method and system for flow controlling | |
CN114640886A (en) | Bandwidth-adaptive audio and video transmission method and device, computer equipment and medium | |
CN102594769B (en) | Method and system for transmitting network media data | |
CN103745682B (en) | A kind of method of asynchronous cascade | |
CN110996120A (en) | Video stream transmitting and receiving method | |
CN106230839B (en) | Receiving control method and device for real-time streaming transmission | |
US7496114B2 (en) | IP image transmission apparatus | |
CN107122320B (en) | Device, system and method for realizing PCM audio playing based on FPGA | |
CN101916219A (en) | Streaming media display platform of on-chip multi-core network processor | |
CN109448738B (en) | Network audio mixing processing method and device | |
JP2005136742A (en) | Data reproducing apparatus | |
CN110445730A (en) | The real-time acquisition and storage method of network data and device based on WinPcap | |
CN115941907A (en) | RTP data packet sending method, system, electronic equipment and storage medium | |
CN112929114B (en) | Low-delay audio distribution method, module and device based on network card | |
CN107040784A (en) | A kind of video buffer processing method, system and device | |
CN105337784B (en) | A kind of loss packet test method, test device and system | |
CN109413485B (en) | Data caching method and device |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
PE01 | Entry into force of the registration of the contract for pledge of patent right | ||
PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: A real-time slicing system and method for streaming media files Effective date of registration: 20220622 Granted publication date: 20180323 Pledgee: Qilu bank Limited by Share Ltd. Ji'nan Zhang Zhuang sub branch Pledgor: SHANDONG TAIXIN ELECTRONICS Co.,Ltd. Registration number: Y2022980008497 |